[Eeglablist] P300 Analysis in STUDY

Joseph Nuamah jknuamah at aggies.ncat.edu
Tue Sep 26 11:50:06 PDT 2017

Hello Tarik,

Many thanks for your email.

I appreciate the pointers you provided.

Kind regards,

On Tue, Sep 26, 2017 at 2:43 PM, Tarik S Bel-Bahar <tarikbelbahar at gmail.com>

> Hi Joseph, some quick responses below, good luck!​
>> *************RESPONSE FOR JOSEPH************************
>> ********************************
>> first ​
>> see the following on google scholar
>> ​ which provides plenty of general guidelines related to your question.​
>> Event-related potentials in clinical research: guidelines for eliciting,
>> recording, and quantifying mismatch negativity, P300, and N400
>> Be sure to ​also review about ~20 recent publications from respected
>> authors using P3 measures in high-impact journals for general guidelines.
>> You may also want to examine reviews about different subcomponents of the
>> P3, such as P3a, P3b, etc... all easy to find on google scholar
>> THEN....
>> 1a. It depends on what your ERP looks like. You should review major
>> methods for computing ERP metrics from various handbooks for ERP/EEG such
>> as the ones from Luck or Handy. See also online tutorials from Luck for
>> ERPLAB that cover various ERP metrics. See also methods in the research
>> articles that you are trying to emulate. You should be able to find at
>> least 20 or 30 articles on Google Scholar, each of which will describe how
>> they computed their P3 ERP metrics.
>> 1b. Use time windows based on published research that has used similar
>> paradigms as yours.
>> Also pick time windows based on what you see in your grand-average ERPs
>> and your single-subject ERPs (the latter will be more variable of course).
>> 1c.​ Yes usually/often midline at frontal, central and central-posterior
>> sites (but depends on a variety of factors).
>> Refer to reviews (such as from Polich or others) on the P3. I think it
>> depends on the particular protocol, type of stimuli, type of population.
>> Reviewing 10 to 20 recent articles (or classic articles) will give you a
>> good enough idea of where to expect P300 dynamics for your paradigm.
>> Anyway, if you just type "p300 EEG topomap" into google images, you can
>> get a good idea of the scalp distribution of p300.
>> ​1d. Get the the time series from the data for a particular channel or
>> group of channels. Determine what part of that time series is your time
>> window of interest. Then average up that data within the time window. You
>> would do so by isolating the timewindow for that channel(s) into a matrix
>> of data, and then taking the average via a matlab function. Of course,
>> please familiarize yourself (Point 1a above) with the various ways to
>> compute ERP metrics.Note also your data should likely be appropriately
>> baselined, etc...
>> 2. See Luck's and ERPLAB's pages on that measure.
>> Just google "half area latency eeglab" and there are several useful links
>> and examples.
>> The blog page that comes up as one of the first links, by Lindeløv,
>> seems kind of useful for your case, which should be translatable to study
>> metrics.
>> Also google "eeglablist and your topic" for some other past eeglablist
>> discussion related to your topic.
> On Mon, Sep 25, 2017 at 6:53 AM, Joseph Nuamah <jknuamah at aggies.ncat.edu>
> wrote:
>> Dear All,
>> In my STUDY design, there are two conditions. I want to, among other
>> things, determine whether the difference between P300 amplitude and latency
>> for both conditions is statistically significant.
>> Please find my questions below:
>> 1a. Given that P300 is a broad component, how do I derive temporal
>> windows for analysis ?
>> 1b. In particular, what time after stimulus onset should that be applied
>> to?
>> 1c. Will that vary across channels (I intend to use  midline sites Fz,
>> Cz, and Pz)?
>> 1d. How do I average amplitudes within these temporal windows for each
>> participant ?
>> For say channel PZ, I am able to retrieve erpdata and erptimes from
>> [STUDY erpdata erptimes] = std_erpplot(STUDY,ALLEEG,'channels',{ 'PZ'});
>> I read about fractional-area (50%) latency in the literature.
>> 2. Can I determine half-area latency with corresponding amplitude for
>> P300 in STUDY?
>> Kindly help.
>> Thanks!
>> Joseph
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
>> sd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170926/af3aed37/attachment.html>

More information about the eeglablist mailing list